Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The attic air tasted of dust and forgotten things. Sunlight sliced through a crack in the boarded-up window, illuminating motes dancing in the gloom. Amelia knelt, her fingers tracing the faded ink on the old, leather-bound journal. It was her mother’s, undisputedly. Inside, the familiar, looping handwriting listed dreams – places visited, experiences yearned for. Beneath "See the Northern Lights," a small, crudely drawn sketch of a green and purple aurora swirled. A tremor ran through her.

She carefully turned the page, the paper thin and brittle under her touch. “Learn to speak Italian.” Her mother, a woman of quiet routine, had never mentioned such a desire. Her breath hitched. Amelia felt a sudden, profound connection, a bridge across the silence of years. This wasn't just a list; it was a map of a soul she hadn't fully known.

Emotion: awestruck

Cluster: Surprise / Confusion
PC1 (Valence): -0.17 Negative
PC2 (Disposition): -1.20

Role in Research

This story is one of 1,000 stories generated for the emotion awestruck. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 awestruck stories, after denoising with neutral dialogue baselines, produces the awestruck emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the awestruck vector is projected through the unembedding matrix.

Promoted:
L0.271
suddenly0.260
stranger0.252
plötzlich0.246
wonder0.242
Suppressed:
la-0.450
싶-0.350
K-0.319
rest-0.248
싶은-0.238